Engineering Safe and Secure Autonomous Systems

This article brings up for discussion aspects which are important for formulating an internationally accepted basic concept for the safety and Secure Autonomous Systems It is also intended to promote cooperation in the international standardization committees (e.g. ISO and IEC).

Introduction

Autonomous Systems powered by Artificial intelligence (AI) are enabling the industry V4.0 revolution. Medical and industrial robots are examples of AS that operate under uncertainties due to dynamic environments, interaction with humans, interaction with other machines.
Ensuring Security and safety is the first step to make solution using in real world with some certificate I ll notice some ones here.

in this article I am trying to :

• Exemplify the need of assured autonomy including safety and security.
• Explain relevant safety and security concepts.
• Select and apply relevant standards for AS.
• Model and analyze Safety-Critical parts of AS.
• Argue the safety and security of AS.

Motivation : Collaborative Robots

Collaborative robots have been redefining human-robot relationships across the manufacturing industry. Companies of all types and sizes have found strategic reasons to acquire or invest in collaborative robots. Therefore,there are many standardization bodies which deal with the safety in human-robot interactions.

Why collaborative robot?

This is a good question before we start, why we need a collaborative robot then we can avoid it and avoid all the risks?

“The ability for robots to sense human activities within its operating environment in real-time and react accordingly becomes a crucial safety requirement.”

Most importantly, the cost of adopting robotic solutions to a given problem has since been reduced significantly with the introduction of collaborative robots.

Flexible production systems capable to deal with: low degree of volume per variant and high degree of product variants.

General idea of collaborative robots: to bring together the advantages of both humans (problem-solving) and robots (precision, repeatability) in one process.

Collaborative Robot : a robot specifically designed for direct interaction with a human within a defined collaborative workspace.

Collaborative Workspace : safeguarded space where the robot and a human can perform tasks simultaneously during automatic operation.

Collaborative Operation (Human-Robot Interaction) : state in which purpose designed robots can safety work in direct cooperation with a human within a defined workspace.

So we move from separate players ===> to team players.

Type of collaborative robots :

Fig. 3. The four forms of collaboration identified by the robot safety standards ISO 10218–1/2

a. Safety-rated monitored stop(SMS): The robot is stopped during the interaction with the operator in the collaboration space.

b. Hand guiding (HG): The safety of the human–robot collaboration is assured by the robot being guided manually and controlled at an appropriately reduced speed.

c. Speed and separation monitoring(SSM): Robot motion only when separation distance above minimum separation distance.

d. Power and force limiting(PFL) : In contact events, robot can only impart limited forces.

The technologies underlying Systems Development focus on the methodological and software aspects used in system development and integration process. This combination of technologies aims to deliver the underlying engineering methods and software tools for robot developers and manufacturers to use during the design, development and integration phases.

The process models are required integrating both:

• stakeholders and their activities over the complete life-cycle (reability for maintenance and development)

we need also to do V&V :

• Verification: have we built the system right ? correct with respect to reference artifact (specification document, Reference system, Model)
• Validation: have we built the right system? adequate for its intended operation?

So after defining a lot of things about collaborative robot, let’s talk now about Safety.

What is Safety?

Safety : Freedom from unacceptable risk of harm. [1]

Risk : is the chance or probability that a person will be harmed or experience an adverse health effect if exposed to a hazard.[2]

A hazard is any source of potential damage, harm or adverse health effects on something or someone.[2]

What is a risk assessment?

The process of detecting hazards and assessing associated risks.

Example: Driving on icy roads is a hazardous condition.

— Risk is lessened by driving slow while using snow tires or chains (less probability of slipping and less damage if you go in the ditch at a slower speed).

— Risk is increased with speed or bald tires (higher probability of slipping and greater damage on impact).

— Risk is zero if you don’t drive at all, even though the hazard still exists! However, if we don’t drive at all, we won’t “get the job done” either!

Risk Management allows us to still do the job, but with the safest (less risk) method.

after that we need to determine the required Performance Level system for safety functions (figure bellow).

To reduce a risk there are not one form like 1+1=2, it depends on the risk and types of hazards.

we can distinguish two safety techniques :

• Passive techniques: that remove risk.
• Active techniques: that control hazards to mitigate risk.

Conclusion :

Next article i’ll talk about techniques and some tools to discuss safety systems. and don’t forget the secure aspect also and some Terminology Safety-relevant, reliability, Tractability, audibility, and some standards ISO 26262 and how to use it in Machine Learning.

Every human being needs to feel safe and secure in their life ‘KETHER DONOHUE’

I think I got it all. As you can see, a lot has changed since the last article i publish. I encourage you to check back on the comments. Some people may either make additions.